where computers learn the law,
and lawyers learn the computer.
In Progress
AI Aces the Bar Exam
April 20, 2026
Evaluating whether frontier AI models can achieve near-perfect scores on the Uniform Bar Examination across the MBE, MEE, and MPT. Currently in progress. Collaborators welcome.
In Progress
AI Aces the Korean Bar Exam
April 20, 2026
Testing whether AI models can achieve near-perfect scores on the Korean Bar Examination, exploring cross-lingual legal reasoning and jurisdiction-specific challenges. Currently in progress. Collaborat...
In Progress
A Survey of Large Language Models in Law
April 20, 2026
A comprehensive survey covering how LLMs are applied in legal reasoning, contract analysis, compliance, and adjudication. Currently in progress. Collaborators welcome.
New
AI Achieves a Perfect LSAT Score
April 11, 2026
Documents a language model achieving a perfect score on the LSAT. Shows that removing internal reasoning reduces accuracy by up to 8 points, and introduces a QLoRA-trained process reward model that improves selection on logical reasoning.
Multi-agent LLM Debate Unveils the Premise Left Unsaid
January 1, 2025
A multi-agent framework that casts implicit premise recovery as a dialogic reasoning task between two LLM agents, achieving state-of-the-art accuracy on premise selection.
Scaling Implicature via Structured Interaction in Multi-Agent LLMs
January 1, 2025
Demonstrates that pragmatic reasoning in LLMs arises not from additional tuning, but from contextualized interaction enabled by a multi-agent framework.
Featured In
